Why do scientific theories become more complex?

Bonini’s paradox is the name given to the problem that emerges when a model of a phenomenon is just as hard to understand as the phenomenon that it is supposed to explain.”

University of Alberta’s Dictionary of Cognitive Science

A critical point of reason here, and one for which we may all be fairly poorly furnished, as much by our intrinsic genetic inheritance or cognitive aptitude as by (all of) our acquired and collective cultural conventions and paradigms of interpretation, is that completeness, wholeness and irreducible endpoints do not have to exist in the world any more than this world has any ultimate obligation to us to be meaningful or intelligible. We live through logic and language and we are endlessly impressed by our own ingenuity in parsing the mathematical syntax and physical grammar of this world into the symbolic abstractions upon which we have built a Global civilisation (such as it is), but the discontinuities and paradoxes underlying this order and aspirational control are rarely, if ever, acknowledged or engaged.

Algorithmic Information Theory (as much as Gödel’s Incompleteness Theorem[s]) demonstrates that there is no shortest, most elementary theory of Global systemic self-containment. We should hardly be surprised that complexity proliferates in the ways it does – the foundational assumptions of our systems of knowledge and science are of an explanatory teleology which has been proven as demonstrably and logically false.

The displacement of intelligibility we experience as we develop more advanced theories and more comprehensive explanations is always going to be in a general direction of complexity as a necessary consequence of the profound interdependencies we encounter. The question is not one of why should sophisticated and comprehensive explanations and descriptions be so difficult to understand, so much as what is it about the nature and presence (by an inverse virtue of their absence) of unity and completeness in anything other than the rank simplest of simulations, theoretical models and organisational systems which provides or guides us towards ways to negotiate this complexity? An absence of unity and complete explanatory systems indicates something profound about complexity, information, logic and, perhaps, also consciousness and life.

Why should the proliferating complexities borne of rational analysis prove to be so unintelligible and cognitively unwieldy? Perhaps because the mechanical, mechanistic logic by which we aspire to ascend to comprehensive or complete explanation can only ever asymptotically approach it’s own underlying discontinuity; displacement into complexity is the information (entropy) cost of these particular machines.